Not sure I'm phrasing this correctly in this context, but I don't think
that service providers are the "rational choice actors" in this scenario
- any more than equipment vendors are.
It seems to me the main "actors" in applying gaming to this problem are
still the end users. It is the case that their actions are limited by
multiple levels of indirection, since end users affect service provider
choices (via service selection), which then affects vendor choices (via
equipment purchases).
In the "actor" mix are equipment vendors (who must justify R&D expenses
in terms of their customer willingness to spend money), service
providers
(who likewise need to justify the amortization of capital expenses and
the expected additional operating costs in terms of their own customer's
willingness to spend money) and end users (whose willingness to spend
money on both services and equipment is somewhat vaguely assumed).
On the flip side, the services that a service provider may offer are
limited by the capabilities of the equipment that vendors provide and
- obviously - the services end users may choose are limited by what's
offered to them by service providers.
I suspect that what makes this hard to use predictively in general is an
entirely subjective "guess" that has to be made with respect the degree
of flexibility the actors have in accepting choices presented to them
by other actors. At what point will end-users choose no services over
any of the service options presented to them? At what point will
service
providers, end-users, or both, choose not to buy any equipment over the
equipment choices presented to them?
--
Eric Gray
Principal Engineer
Ericsson
-----Original Message-----
From: Hallam-Baker, Phillip [mailto:pbaker(_at_)verisign(_dot_)com]
Sent: Thursday, March 15, 2007 11:58 AM
To: Tim Chown; ietf(_at_)ietf(_dot_)org
Subject: RE: Game theory and IPv4 to IPv6
So the rational choice actors here are the ISPs not the end-users.
Build that constraint into the model.
-----Original Message-----
From: Tim Chown [mailto:tjc(_at_)ecs(_dot_)soton(_dot_)ac(_dot_)uk]
Sent: Thursday, March 15, 2007 10:53 AM
To: ietf(_at_)ietf(_dot_)org
Subject: Re: Game theory and IPv4 to IPv6
On Thu, Mar 15, 2007 at 07:37:26AM -0700, Hallam-Baker,
Phillip wrote:
The problem is that until IPv6 has critical mass it is much
better to be on IPv4 than IPv6.
If there are any grad students reading the list take a look
at the game theory literature and apply it to the transition.
Assume that it's a rat-choice world and that each actor
follows their best interest.
An actor can be in one of several states:
Unconnected
IPv4 connected with own address
IPv4-NAT connected with NAT address
IPv4/IPv6 connected Dual stack
IPv4-NAT/IPv6 connected Dual stack
IPv6 connected
Unfortunately most of the rats cannot choose certain states,
so the game
is fundamentally flawed. The ISPs are keeping the cheese to
themselves.
Squeak.
--
Tim
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf